*****I Still think there's a way to make electronic voting work*****

[PretzelWarrior]

I was reading Greg Palast piece in the Nation and kind of sketching out the various scenarios. The ONLY scenarios where I think it makes sense in ANY mass counting type of nomenclature is to have periodic auditing of the vote count by outside nonpartisan agencies both in the after hours for "early voting" days and for specific machines during election day.

I think it is a purposeful problem willfully created by leadership in BOTH parties.

I believe an electronic voting machine that spits out the voting results for you, the voter, and another half of the receipt acting as a manual recount and audit ballot would be entirely doable.

I also don't understand why they don't mandate ONE F-ING SYSTEM for all states and all precincts. If I vote in fairly affluent suburb of Portland, Oregon it should be the same as someone in downtown Philadelphia or in Podunk, Arkansas voting.

I have decided the entire REGISTRATION PROCESS and mechanics as well as the entire VOTING PROCESS and mechanics needs a good overhaul again. We certainly have enough variations on test scenarios to say which has performed the most reliably.

Any thoughts? I seriously believe this needs to be something driven by us, the Obama supporters and other grass roots activists who care that we have such a piss poor voting process in the richest and most innovative country in the world that supposedly cares about democracy.

[dbonds]

1. It CAN work, but it has to be open source software and system design.

Everything would be peer reviewed to make it hack proof. And it needs to print a paper receipt.

[PretzelWarrior]

2. exactly. I was actually thinking about audits

and a 2 piece detachable receipt on cardstock. one half for you the voter. the other half for the elections board to retain.

I agree about open source and chain of custody and tests on machines with printouts that are saved.

[PretzelWarrior]

7. note to self. part of the issue is the horrid GUI and "save" features.

it should have a great navigation view allowing a person to go up and down the ballot and get to a particular choice with very clear voting choices. as a person goes from screen to screen it should be autosaving accumulated information thus far.

once it prints your filled out ballot and you get to a screen where you agree all choices have been selected (with maybe any items you did not make choices for highlighted) you confirm it and THEN it gets recorded as a vote.

There are so many ways. If the voting machines were the same everywhere (you know, like the stop signs are the same everywhere and mailboxes are the same everywhere) they could have voting handbooks available online and hard copy. You could have tutorials for how to navigate in a voting screen well before you get to the polling place.

and this doesn't even get into the improvements that need to be made in voter registration being a straightforward process.

Am I dumb? It seems like to register you should have to provide SOME type of proof of who you are. If in that state you DON"T have to...then shouldn't you have to show your ID at the polling place?

Maybe I'm oversimplifying this. But right now I'm pissed about all of this hurdle to new voters getting to vote.

[annabanana]

22. paper receipts, if they are not the legal instrument of a BALLOT are useless

The ballot is what gives them the right to govern us on our behalf. .

If order to use even the most secure computer in the election process, , there would have to be a mandatory recount sufficient to supply statistical certitude of accuracy.

We'd be better off using a pencil and paper and hand counting. If India can do it why can't we?

[Tigress DEM]

3. 1 suggestion I thought made sense. Touch screen prints out ballot marked, voter can review in hand..

THEN when voter verifies ballot matches voter can confirm touch screen, but the ballot that is COUNTED is the in hand paper ballot the voter has seen with their own eyes is marked correctly. THAT is put into a secure optical scanner and the running tallies from the scanner and touch screens are compared randomly during the day.

[lutherj]

4. I've had the same thought. In fact, the scanner can be made by a different vendor

or made to use open source code as suggested above. If the tally from the touch screens fail to match the tally from the scanner, then the paper print outs can be counted by hand.

[PretzelWarrior]

5. I like that idea. It is interesting. it would require that specific machines

be printing out ballots that trace back to specific machines. that way the comparison is valid right down to each machine.

God, we're a bunch of geniuses. ;)

But seriously, these are some great ideas.

[Tigress DEM]

6. It was another DUers idea, but I modified the trace back idea. Techie smart not Genius. nt

[RichardRay]

14. It currently exists in prototype, and it's not expensive.

For a good overview, see:

Citizen-Verified Voting (CVV)

The same algorithm is used to provide secure processing of financial instruments.

CVV provides:

Integrity – Correct vote count.
Anonymity – I can’t tell how you voted.
Involuntary Privacy – You can’t prove to me how you voted.
Voter Verifiability – You, the voter, can verify the first two goals.
Public Verifiability – Anyone can verify the first three goals.
Robustness – If something goes wrong it can be detected and fixed

The system generates an encrypted receipt that contains no information unless it's matched with the ballot to which it's paired. In order to be provably secure the system requires that some 'non-negligible' number of voters choose to verify their ballots.

Item #3 above is the one that no duplicate ballot type receipt can provide. It provides security against using the ballot system to verify a 'vote buying' scam.

The whole this isn't hard. It does take some understanding of data encryption to understand how it works, but it's not rocket science.

[PretzelWarrior]

15. I like that idea. I hadn't thought about how #3 was important

so I'll check out that website. I love it that people are thinking about this. Now we just need to get the political will to fix this.

[Political Heretic]

8. Can we stop with the ****** on everything already? :) ***** oh yeah, and * for good measure.

[PretzelWarrior]

11. NO, we can't. I can't stand when my post sinks like a stone

and it is an important topic but no one pays attention to it if it says, "Thoughtful introspection on an important issue".

Sorry you have a problem with it. :) just teasing you.

[Political Heretic]

16. Then write posts that your readers deem valuable, and dial your ego down to 5

..would be my advice. :)

[PretzelWarrior]

17. uhhh....you're not smart

if the title doesn't have something to get people to read it....then it won't even be clicked on and replied to which means it will sink like a stone.

Thanks for the great advice. My ego is now only 100,000 candle power thanks to your smiley.

[frickaline]

9. we need to stop fearing this technology and start FIXING it because it is here to stay

I agree with you, we can make this work if we try. Running away from electronic voting isn't the right answer. Engineering a better solution, one that really is safe and reliable is the path we should be on. We are progressives after all! :)

The poster who mentioned open source code is absolutely right. Receipts are a must as is a confirmation number that could be used to verify your vote. Vote tallying should be made public (over the web). You should be able to see each individual vote and associated confirmation number as part of the overall vote count for an area. This way, every person could visually verify that their own vote was individually counted, which is something we cannot do with paper ballots.

[PretzelWarrior]

12. I had pondered that after the last election. VERIFICATION

I mean, isn't there some way you could use an ID number that you alone know to look up your results online?

I decided that probably isn't necessary when I looked at other things available to us. But all of the ideas out there vs. WHAT WE REALLY GOT shows me they are purposefully acting as obstacles to a superior solution over there in Congress.

[eridani]

10. Even if it was reliable, DRE voting is still an expensive, stupid disenfranchisement mechanism

It conflates the very slow act of human voting with the fast act of tabulation. I timed one voter (opscan paper ballot in a privacy booth) during a slow period in the primary at 13 fucking minutes! Had he been at a DRE during lunch hour, he could have easily disenfranchised a couple of dozen people who had to get back to work. 10 or 20 voters can simultaneously be filling out paper ballots. It takes a second for the precinct opscan to read them. If they didn't fill in the ovals correctly or left stray marks, they can try again with another ballot. If the precinct optical scanner breaks down, the ballots can be collected for later tabulation by the central scanners used for absentee ballots.

And of course there must be mandatory hand count auditing in all elections, not just close ones.

You have to look at the whole process, not just the technical parts of it.

[PretzelWarrior]

13. I agree with you in some aspects

there were also plenty of voters who found out how stupid ballots make it difficult to know you are voting for the one you wanted.

I think if GUI is set up correctly it can make voting choices very apparent and can ensure who the person chose.

THis is why I'm getting this conversation going. Because I think there are still process issues around the memory and tabulation software/hardware in optical scan, that needs to be looked at too.

I can tell you that if EVERY state used the same exact open source code electronic voting machines, the cost of them would go down. In addition, by measuring the typical voter participation rates and doing other work ahead, the states can decide where voting machines should be in greater supply and where they can have less. By timing typical time of transaction (not outliers like one woman who took 13 minutes) the elections officials can get close to something optimal.

[eridani]

20. GUI setup is entirely unrelated to the basic problem here

Human voting is SLOW, sometimes very slow. Tabulation takes milliseconds. Why chain the second process to the first? To rearrange the old saw--if it's not worth doing at all, it's not worth doing well. Opscan puts large numbers of the slow processes in parallel to the fast process of tabulation. DRE forces those processes into a highly inefficient series. No conceivable design can alter that basic fact.

Also, measuring typical voter rates is useless. There is still the issue of having to have many more machines for peak hours before and after work and during lunch hour than during slower periods. Also there is the issue of crashes, malfunctions and power failure. No problem with opscan--just collect the ballots and tabulate them later.

[PretzelWarrior]

21. I see what you're saying, and you may have some good points...

but I still think if you go that route you will need to state that optical scan ballots are at root an electronic voting problem. This means the traceability has to be addressed.

I kind of disagree with you that it takes too much effort to get enough working machines in a polling location to do the job and have votes quickly tabulated and cross checked to ensure the vote is reliable.

I'm still open to weighing all the options, because I want to jot down the pros and cons for all of the various methods.

[eridani]

23. Yes, but optically scanned ballots can be audited by hand counting

--which IMO ought to be mandatory. I'm all for open source, but as a lab chemist I find a set of standard weights more useful for checking the performance of my scales than a set of schematics detailing the scales' construction. First things first. We need to develop no new technology to do hand count auditing.

[Poseidan]

18. screw electronic voting

If they want to invest in new technologies, they should go with internet voting.

[PretzelWarrior]

19. are you kidding? P.S. internet voting is also a form of electronic voting